EMG Signal Processing During Whole-Body Vibration: A Preliminary Study

International audienceEMG SIGNAL PROCESSING DURING WHOLE-BODY VIBRATION: A PRELIMINARY STUDY University of Nice-Sophia Antipolis, Nice France 1:I3S; 2:LAMHESS, Faculty of Sports Sciences Introduction Whole-body vibration (WBV) has become one of the most popular alternative exercise modality. To investigate the mechanisms leading to increased neuromuscular performance following WBV exercises, numerous studies have examined the neuromuscular adaptations by recording surface electromyography signal (sEMG). However, sEMG during WBV is problematic since it contains sharp peaks at the vibration frequency and its multiple harmonics. So far, different approaches have been conducted to cut out the excessive sEMG bursts which consist of exclusion of a wide band-width frequency (Hazell et al. 2007) and elimination of the signal at the vibration frequency and its harmonics with the help of band-stop filters (Abercromby et al. 2007) or interpolation technique (Wakeling et al. 2002). Hence, there is no scientific consensus about the most adequate method to process sEMG during WBV. Another issue to be addressed is the sEMG analysis after the filtering process. For instance, the sEMG has been compared between sham and WBV conditions and/or normalized to the sEMG measured during a maximal voluntary contraction (MVC). However, to our knowledge, the normalization of the sEMG to the electrophysiological response (i.e., M-wave) evoked by nerve stimulation has never been done for WBV. Therefore, the aim of this study was to compare: 1) sEMG processing methods during WBV, namely band-stop filters vs. the interpolation technique and 2) sEMG normalization procedures (i.e., sEMG of a MVC vs. M-wave). Methods 15 physical education students performed isometric semi-squats during sham and WBV conditions (4 frequencies and two amplitudes) on a vertically vibrating platform while measuring the sEMG of the quadriceps femoris muscles. Femoral nerve electrical stimulation served for the assessment of M-wave responses. MVC of the knee extensor muscles were also performed. sEMGrms values were computed in the frequency domain after signal processing by using a band-stop filter or interpolation technique. These sEMGrms values were then normalized either to the sEMGrms value obtained during an MVC or to the respective muscle M-wave. Results We are currently colleting the data of this experiment. Due to the wide variability of results in the literature, it is hazardous to formulate a hypothesis. However, since the M-wave is the electrical equivalent of the recruitment of all motor units, we assumed that the normalization of the sEMG to this parameter will provide a better method and thus improve the scientific knowledge related to WBV exercise. References: Abercromby et al. (2007). Med Sci Sports Exerc 39:1642-1650. Hazell et al. (2007). Appl.Physiol.Nutr.Metab. 32,1156-1163. Wakeling et al. (2002). J. Appl. Physiol. 93:1093-1103